What is statistical significance?

A way to have assurance that experiment results are entirely accurate and reliable

A method for mathematically demonstrating the reliability of a given metric

A process for calculating whether an experiment is valid as an integration on your website or app

The percentage of statistical uplift you can expect as the result of an experiment
Which of these statements is true?

An experiment can’t end unless statistical significance is reached

Your chosen level of statistical significance reflects how comfortable you are with risk.
Miranda has to run an experiment, but due to a tight deadline, she decides to set the significance to end it at 80% instead of the 95% statistical significance they usually aim for. What risk is she running in doing this?

She’s taking a 15% higher risk than usual that the experiment results aren’t accurate.

There’s only an 80% possibility that the information gathered so far reflects the activity of users of the site in the experiment period.

She has a 15% chance that her results won’t be as accurate as they usually would be.

At 80% significance, there’s really no significant risk that her results are inaccurate. She can be just as confident in her results as she usually is.